A 2.0 Analysis A2.1Performance of the Steering Control and Steering Gear A2.1.1 Sequence of Events According to the DGPS replay from the pilot's laptop, a decreasing trend in vessel speed had begun at 1438:10. This indicates that the turn to starboard was initiated; the drag force created by the rudder and the increased water resistance on the vessel's port side began to be felt. A vessel making a turn with a set rudder angle will reach a maximum rate of turn for that angle and then remain stable at that rate until the rudder angle is changed. The time that the rate of turn remains stable can be considered a close approximation to the time the rudder remains at the set angle. Simulations conducted using maximum rudder angle at varying time intervals illustrate this result.20 The rate of change of COG was analysed via data from the DGPS replay. Since wind and current were near constant, and the time interval brief, the rate of change of COG can be used here as a close analogy for rate of turn. This information shows a maximum rate of change to starboard was maintained for a very short time (15to20seconds) before it decreased quickly to near zero. These data corroborate the sequence of events as reconstructed from observations of those on the bridge, in the engine control room and in the steering flat - in particular, the 15-to 20-second interval during which the helm did not respond and remained hard-to-starboard, after which the helm was placed at midships and the rudder responded. The data collected thus indicate that the rudder remained to starboard, at varying angles, for approximately 40to 50seconds, of which 20seconds were at the hard-to-starboard position (35) and after which it went to midships. The rudder did not respond to helm orders for about 15to 20seconds. A 2.1.2 Accidental Deactivation of the Ship's Wheel The ship's records show no intermittent or chronic problems or any unusual maintenance on the steering gear or its components. Extensive testing of all electrical components (steering station, wiring and components at the steering gear), on-board inspections and extensive bench testing of the hydraulic components of the steering gear were carried out; no malfunction was apparent. Although the flow capacity of the steering gear's hydraulic pumps had diminished about 17% over 22years of service, the steering gear could still produce a theoretical torque in the range of 40t/m. Flow rates measured during bench testing would produce a torque of 32t/m at an operating pressure of 103.5bar. This would have been more than sufficient to execute rudder movements with the vessel making 12knots through the water. The calculated torque necessary to hold or move the rudder under these conditions would have been about 12.65t/m. No intermittent failure was reproducible during extensive operational tests after the grounding, and there was no report or record of any intermittent failure in the past. It is unlikely that there was an intermittent failure of the steering gear at this exact time, and for this duration. Given the length of time the helm did not respond and the reported return of control after the OOW had toggled the SMS switch, loss of rudder control was probably due to the SMS switch being at a position other than HAND. Photo1. Simulation of accidental switch activation The SMS switch of the helm unit is 55mm behind the spokes of the hand steering wheel. Simulations conducted after the accident revealed that the switch could be accidentally thrown right, to the NFU position, when the wheel was turned to the right, or left, to the OFF position, when the wheel was turned to the left.21 With the low detent torque of 0.6Nm, this switch was relatively easy to turn by simply leaving the fingers of the hand extended while turning the wheel (seePhotograph1). The location and orientation of this switch were such that accidental activation was possible. There was no protective barrier or alarm for this switch. Accidents in various transportation modes have been attributed to a less-than-adequate location and/or design of safety-significant controls.22 Control actuators should be designed and located so they are not susceptible to accidental activation. Methods to reduce the likelihood of accidental activation include: locate and orient the control actuator so that unwanted activation is unlikely; provide sufficient control resistance (detent torque) to prevent unintentional movements; require complex motions for control activation such as an interlock or rotary motion; and isolate controls or provide some sort of physical barrier.23 Many models of ship helm units incorporate safety features, such as alarms or guards on the SMS switch. These switches are also in a location and orientation that reduce accidental activation. Also, the detent torque on many models has been observed to be considerably higher than it was on the Alcor. It is not possible to determine with certainty whether the SMS switch was moved to the NFU position (while the pilot was turning the wheel to the right) or to the OFF position (while the pilot was turning the wheel to the left). Either event would have resulted in wheel deactivation. The movement of the switch from HAND to either one of these locations would explain the brief lack of response from the rudder just prior to the grounding. A 2.2 Ship-handling and Pilotage Experience The vessel's head was coming to starboard even before the pilot assumed the helm. Even at a rate of turn of one third of a degree per second (a very slow turn), the 9.5 starboard alteration would not have taken more than 30seconds, leaving two and a half minutes to position the vessel to the starboard side of the channel before meeting the downbound vessel. Instead of letting the vessel react in the time required, the pilot attempted a hard-over/hard-over counter-rudder manoeuvre. The Alcor had a static depth of water to draught ratio (Dw/d) of 1.67 while off buoy K-108. If squat is considered, this ratio approaches1.5. Since becoming a ClassC, the pilot's experience on vessels (other than the Alcor) with reduced under-keel clearance, and thus with reduced Dw/d ratios, was very limited. A ship can experience shallow water effect when the depth of water is less than twice the draught.24 Shallow water effect becomes significant when the ratio of Dw/d is equal to 1.5. When this ratio is 1.2 or less, full shallow water effect is felt.25 Full shallow water effect can double a vessel's deepwater turning diameter.26 From Les Escoumins pilot station to Cap Gribane, the channel is wide and deep and vessels there display normal, deepwater manoeuvring characteristics. In the Traverse du Nord, however, the channel narrows to 305m and is shallower. Additionally, at buoy K-108, the channel is dredged to keep the minimum depth at 12.5m, and the 10m depth contour is close to either side of the channel. In contrast, at the other course alteration in the Traverse du Nord, the channel is deeper and the deep water extends well beyond the channel limits. The Alcorwould have experienced the following effects and forces during the starboard turn at buoy K-108: shallow water effect - an increased resistance on the port side abaft the pivot point and the shifting aft of the pivot point due to increased resistance forward and on the port bow due to the low Dw/d ratio of 1.67 (this ratio is further reduced, to 1.5, if squat is considered);27 approximately 0.89 m of sinkage due to squat - being a large, full-bodied vessel, the squat tends to be more by the bow than by the stern, thereby adding to the 7 cm static forward trim, further decreasing rudder efficiency; and a tendency to go bodily to starboard due to a two to three knot following current, approximately 75 abaft the port beam. The cumulative effects of the above would reduce rudder effectiveness and would account for the sluggish behaviour and resistance to the starboard course alteration at this point. Although the helmsman executed all helm orders correctly and without hesitation, the pilot opted to assume the helm himself at this juncture. Given the developing situation, the presence of another vessel 1.5 nm away, and the actions of the pilot at this time, this action suggests that the pilot felt pressure to get the Alcorturned, to remain on the starboard side of the channel. This is further reflected in the pilot applying full helm for a course alteration of 9.5. Efficient and effective ship-handling requires thinking ahead of your ship so the vessel reacts to helm and engine orders, and not to make helm and engine orders as a function of the ship's movement. Additionally, as a vessel's DWT increases, so does its momentum, which decreases its ease of handling, especially in confined waters with low under-keel clearance. The larger the ship, the longer she takes to respond to helm and engines; the more judgements must be based on advance knowledge and not on observation alone.28 The pilot's actions suggest that he did not fully appreciate or anticipate the various adverse effects on the vessel's manoeuvrability due to the low Dw/d ratio, squat, and following current. A2.3 Pilotage: Training, Experience and Risk-based Methodology A typical risk assessment methodology takes into consideration, among others, the following: identify the problem and associated risk factors, and develop an information base related to the risk factors; form a risk management team to carry out the risk assessment; identify and consult with all interested parties and determine their risk concerns; analyse risk scenarios and their frequency, consequences, and cost implications, as well as interested parties' acceptance of risk; identify risk control options and their effectiveness and cost implications; assess interested parties' acceptance of proposed actions and residual risks; and establish a process to monitor the chosen action. The need for such an approach has been identified in the 1999 Canadian Transportation Agency (CTA) Pilotage Review Report, and is reflected in recommendation1, dealing with designated compulsory pilotage areas. As a result of changes to the licence class required for a particular vessel size, the required pilotage experience (excluding apprenticeship) of a licensed pilot on a ship of more than 25000DWT (such as the Alcor, at27536DWT) has diminished as follows: Under the current regulation / service agreement combination, a pilot is permitted to work on ships up to 30000DWT one year after becoming a pilot. This level of experience (excluding apprenticeship) is one quarter of the level of experience that was required on vessels greater than 25000DWT before July1997. A2.3.1 Pilotage Training, Experience and Performance Measurement A 2.3.1.1 Training Requirement The two-year pilot apprenticeship program for District2 covers technical and local knowledge aspects of the profession and provides hands-on training. Apprentice pilots may select the vessels they wish to work on and are encouraged to choose as wide a variety of vessels as possible. Experience on deep draught vessels during apprenticeship is required (per the pilot corporation's training plan) to the extent of three trips per year on vessels with draughts of 12.8m or more. However, this accounts for only 2.5% of the specified number of compulsory trips per year. Although information obtained from the pilot corporation and the pilotage authority indicates a higher level of exposure to deep draught vessels than these minimum requirements, no formal system is in place to measure and evaluate the apprentice's capacity on these vessels, and in particular the candidate's ability to handle vessels having low under-keel clearance. With changes in the pilot licensing regime, pilots are now required to handle larger tonnage vessels earlier in their pilotage career without a requirement to gain experience on these vessels and/or through a specialized ship-handling course using manned model training.29 A 2.3.1.2 Ship-handling Experience and Training Ship-handling skills are acquired through a combination of formal training and practical experience. The pilot must acquire sufficient knowledge to continually evaluate the navigational situation in order to make decisions and/or take appropriate measures to safely pilot the vessel. Formal training reinforces the knowledge of basic concepts. However, emphasis on practical experience is paramount, as it provides an opportunity for a pilot to apply those concepts in varying operational circumstances. In the past, the longer time spent on smaller vessels allowed progressive accumulation of ship-handling experience as a pilot progressed through his/her career path. A small vessel, being more readily recovered from a ship-handling error, can add to the experience base of a pilot while the potential for adverse consequences is minimized. This slower progression to larger vessels provided pilots with an opportunity to deepen their experience base before serving on larger vessels. Realizing that change was necessary, an initiative to adjust to larger ships navigating the St.Lawrence saw the minimum level of certification for pilotage candidates raised to Master Home Trade (or First Mate Foreign Going). While it is recognized that there is an increase in sea time for persons seeking these certificates, this does not assure that a candidate will be more skilful in ship-handling. Specialized ship-handling courses, designed to accelerate the experience base of pilots, can be used to supplement the knowledge gained through apprenticeship. If given early enough, such a course would prepare pilots to handle larger ships at each stage of their license progression. A 2.3.1.3 Current Practice Recent reductions in the experience base for licensed pilots on 25000 to 30000DWT vessels have not been offset by earlier training (such as the ship-handling course) or clearly defined criteria in the apprenticeship program that places a greater emphasis on vessels with low under-keel clearance. Further, under the current regime, pilots progress from one class to another, based strictly on completion of a minimum number of assignments in the required time. The ability to handle larger vessels is not qualitatively assessed at each or any stage. Current practice within District2 calls for all new ClassB pilots to be sent for ship-handling training during their first year. With LPA regulatory/ service agreement changes, the C-1class is now the equivalent of the old B-3class. These pilots have less experience than their B-3 predecessors but they are not offered the same training until they become ClassB pilots. Given that pilots are required to work on larger vessels earlier than before, the extension of such training to the C-1pilots could help increase the training/experience and better prepare pilots for larger vessels. A 2.3.1.4 Pilotage and Safety The primary purpose of pilotage is safety. Compulsory pilotage areas are established for the benefit of the community - to protect the environment and port infrastructure from marine accidents. There is an expectation that a pilot's performance and operational procedures are of a standard that is internationally recognized and accepted. The Board, recognizing the need to maintain the highest practical safety standard for vessels operating in Canadian pilotage waters, recommended that pilotage authorities develop and implement a safety management quality assurance system.30 The Minister of Transport accepted the recommendation and tasked the pilotage authorities to develop a pilot quality assurance system. This has also been addressed in recommendation9 of the 1999CTA report, which reads, in part: that the pilotage authorities be required to develop and implement a fair and reasonable system for assessing pilots' competence and quality of service, after consultation with interested parties. This assessment process should take place regularly and not less than every five years. Pilots working on larger ships earlier in their careers has increased the need for an effective quality assurance program. Given the reduced experience base, a competency-based training and evaluation program will permit objective evaluation of a pilot's abilities to safely navigate a vessel. A 3.0 Conclusions A 3.1 Findings as to Causes and Contributing Factors The combined effects of squat, shallow water effect and a following current, in conjunction with the vessel's speed and the low under-keel clearance, contributed to the sluggish steering behaviour of the Alcor off buoy K-108. The pilot's experience and training was such that he did not fully appreciate or anticipate the undermining effects of low under-keel clearance on the vessel's performance. By assuming the helm and employing hard-over wheel for a minor course alteration, the pilot, perceiving an emergency situation where none existed prior to his assuming the helm, set in motion the chain of events that resulted in the grounding. The location, orientation, and low detent torque of the steering mode selector switch, and the absence of a mechanical guard, probably allowed for the accidental deactivation of the ship's wheel, which would account for the brief lack of response from the rudder (15to20seconds) just prior to the grounding. An alarm would have permitted the movement of this switch to be noticed by the navigation team at this critical time. A 3.2 Findings Related to Risks The LPA allowed incremental regulatory and service agreement changes to go forward without the benefit of a formal risk assessment. This permitted handling of larger vessels by pilots who may not have been fully prepared to do so. The current apprenticeship and post-apprenticeship training program does not qualitatively evaluate a candidate's ability to handle larger vessels or vessels with low under-keel clearance. A 3.3 Other Findings Flow capacity of the steering gear's hydraulic pumps was found to be diminished by approximately 17%, compared to original specifications, but was sufficient to execute rudder movements. B 2.0 Analysis B2.1 Contingency Plans and Risk Assessment In the hours following the grounding, the onus to make the right decisions as to refloating was on the master, with the assistance of the pilot. Because the vessel was not initially hard aground, but still moving, attempts were made to free the vessel under her own power. Only one tug was ordered approximately 1.5hours after the grounding, after permission was sought from the vessel's owners. Valuable time was lost, as by this time the vessel had moved higher onto the shoal with the rising tide. When the tug arrived, the tide had dropped about one metre. One tug was not sufficient to refloat the vessel and it sustained heavy damage before it could be refloated. The company ISM Code procedures contained a generic checklist of actions to be taken in situations such as groundings and salvage. As each event occurred, the master, pilot and various governmental authorities reacted to events rather than take action based on a structured approach, such as could be procured from contingency plans or another risk management process. Under the ISM Code, the safety management system clearly establishes that the master has overriding authority and responsibility to make decisions regarding safety and pollution prevention. In addition, the Protection and Indemnity Clubs recognize the urgency attached to emergency situations, and protect an owner against the financial consequences associated with the need for the master to make decisions without the benefit of cost estimates. In reality, however, the master is under pressure from the owners, either implicitly or explicitly, to consult with the company and keep expenses to a minimum under any circumstances. This is reflected in the master consulting the owners about tugs subsequent to the grounding of the Alcor.More often than not, time zone differences between the vessel and the owners' place of business, and language barriers that may exist between the master and owners, can make communication between them less than optimal and may be onerous for the master at a critical time. Additionally, the master is on location, and so is best suited to evaluate and make decisions regarding the quantity and quality of resources, rather than owners who may not be aware of all pertinent local factors. The master, however, does not necessarily possess in-depth knowledge of the grounding area or the resources available, and must rely on local authorities for guidance. The absence of local procedures, contingency plans or risk assessments (TC, LPA, CCG) can cause delays, and selection of appropriate resources for the task is not optimized. A prompt refloating is critical to reduce risks to the environment and to the structural integrity of the vessel in high-risk waters, such as in the vicinity of the Traverse du Nord. The waters between Qubec and the Les Escoumins pilot station can, in many ways, be considered high-risk waters. Hazards include: dense fog, strong currents, high winds, dense pleasure craft activity and whale watching during summer months, fewer floating aides to navigation during the winter months, high tidal amplitudes (as much as 7.1m off le-aux-Coudres), heavy ice formation, areas with low under-keel clearance, and deeply laden vessels that regularly transit on top of high water only -(Traverse du Nord). In the event of a grounding, the level of risk associated with vessel and environmental damage varies from the time the vessel has grounded until the vessel is safe at her berth. Prompt summoning and dispatch of resources, concurrent with the initial attempt to refloat the vessel, is essential to maximize her chances of refloating. A pilot can be a valuable resource to the master during refloating. LPA has no structured approach to prepare pilots to help masters assess risks and make informed decisions to respond to emergencies, decisions that could have an impact on timeliness and appropriateness of resources. CCG has a coordinated and structured emergency management approach, using a risk management model, for emergencies associated with pollution, search and rescue activities, and harbour operations. A similar approach, however, is not used for navigation-related emergencies (such as groundings) for vessels transiting narrow channels and in pilotage waters. It is important to understand the risk profile of a port or waterway in order to establish risk management priorities....34 The dynamic nature of salvage operations often requires ad hoc problem solving and decision making, a situation conducive to increased error. A structured approach, therefore, provides a framework around which informed decisions can be made. Also, it helps all parties to coordinate, communicate and closely monitor the developing situation. Legislation allows for timely action in responding to shipping emergencies. The Canada Shipping Act (CSA) permits the Minister of Fisheries and Oceans to take appropriate measures to minimize and prevent pollution damage, with no limiting time criterion.35 The Navigable Waters Protection Act allows action to be taken by the Minister of Fisheries and Oceans if the obstruction or danger continues for more than 24hours.36 Given the tidal amplitudes between Les Escoumins and Qubec, the latter may be too long a period under certain circumstances. B 2.2 Factors Affecting Salvage Operations The knowledge and experience base necessary for salvage operations and pilot/salvor interaction are both important contributing factors to the success of a salvage effort. Failed salvage efforts, where these factors have been less than adequate, have been noted.37 A complete salvage plan includes preparation of the vessel for refloating and navigation of the vessel once afloat. LPA and the Pilot Corporation, with in-depth knowledge of local waters, are best suited to provide input regarding local navigation. Their participation in the planning and development phase would help ensure, among other things, that: the navigation plan is precise and complete, local factors critical to the success of the mission are considered during the planning phase, the role of all participants is clearly understood, the stage at which conduct of the vessel will change hands from the salvage master to the pilot is clearly identified, and effective use is made of all available resources and technology. Salvage operations are complex, dynamic operations that require good teamwork and coordination among the salvage master, the salvage captain, the pilots, and the master of the vessel. These can be achieved through effective communication and continuous monitoring of the evolving situation. Another contributing factor to the success of a salvage operation is implementation of a developed salvage plan; a plan with which all members of the team are conversant. B 2.2.1 Communication and Monitoring Lack of communication has been identified as a factor in a number of marine occurrences.38 During both refloating attempts, communication among members of the bridge team respecting navigation of the Alcor was limited and casual. Consequently, individuals on the team were not fully aware of the developing situation. Reduced awareness resulted in the bridge team's uncoordinated, improvised and untimely response to the evolving situation. Also, critical decisions that could have affected the refloating attempt, such as closing/reopening of the channel, were not communicated to the team.39 Lack of communication between pilot(s) and other bridge team members fosters a fragmented working relationship with consequential breakdown in the synergy of the team. This method of decision making and communication can then become a weak link and the system becomes prone to single-point failure. B 2.2.2 Effectiveness of the Team When manoeuvring in confined waters, all of a vessel's navigation equipment, including radar, must be used to advantage, and pertinent navigation and vessel safety information must be effectively communicated to optimize and maintain situational awareness among the bridge team. During both the failed and successful refloatings: the navigation team did not function as a cohesive team, in that information essential for maintaining situational awareness was not communicated to team members;40 positions were not plotted on the chart and navigation equipment was not used to advantage, in that radar techniques that would have helped team members to visualize the manoeuvring room astern and around the vessel were not used effectively; the manoeuvre was such that it permitted the vessel to remain abeam the current, and action to counter the current was ineffective - this despite the presence of four tugs; and none of the navigation team members fully appreciated that the vessel was setting onto the shoal. This would suggest that the salvage/navigation team did not fully appreciate the effect of the current on the vessel and that the progress of the vessel was not closely monitored.41 This culminated in a second grounding (10November1999) and in a narrowly averted third grounding (05December1999). B 2.2.3 Precision of the Navigation/Refloating Plan The plan (successful refloating, 05December1999) for the navigation of the vessel once afloat was reviewed and found to be incomplete: The con was passed too early, which resulted in the pilot improvising manoeuvres for re-entry into the channel. A dedicated helmsman was not posted at the time of the vessel's refloating. This resulted in members of the bridge team having to activate the helm on an as needed basis. The closing of the channel was improvised. This was done by TC officials, but thereupon not adequately communicated to other members of the team. The pilot's decision to permit another vessel to transit the area at the time of refloating came into conflict with this decision. The reopening of the channel was tacit and premature. Decisions by pilots and lack of traffic control on the part of VTS resulted in confusion and a disorderly flow of traffic and contributed to a near-collision between a tanker and a container ship. (SeePartC of this report for more detailed facts and analysis on this related event.) B 2.2.4 Deployment of Resources B 2.2.4.1 Electronic Chart Systems The Starlink tracking system (utilizing differential global positioning system technology) used by the pilot on 09 November 1999 was designed for use in the navigable channel only, and was not employed during the first refloating attempt on 10November1999. However, the precise real-time vessel position, course over ground (COG) (withvisualvector) and speed readouts produced by Starlink would have been a valuable asset in increasing the situational awareness of the team. Even without a chart overlay capability, this system could have been used, in conjunction with radar information, to help the bridge team monitor the vessel's progress. The vessel was not equipped with an electronic chart system (ECS), nor was one required by regulation. During the second salvage operation, a portable unit with ECS capabilities was not brought on board by the salvage team; nor was a Starlink tracking system used by the pilots to assist in maintaining situational awareness. These systems are capable of providing a continuous, precise update of the position, along with the vessel's image (to scale).42 This, together with the visual representation of the forces of wind and current acting on the vessel, provides valuable information that would aid the situational awareness of the bridge team. Additionally, during the successful refloating, an electronic chart display information system (ECS/ECDIS) would have provided precise COG information that could have compensated, to some extent, for the absence of accurate heading information from either the gyro compass, or magnetic compass. The benefits of using an ECS/ECDIS for salvage operations in restricted waterways were not fully appreciated. B 2.2.4.2 Tugs In this instance, four tugs were used to refloat the vessel. Tugs secured on a line forward and/or aft were used to turn the vessel and for speed control. The remaining tugs were left unsecured to provide mobility and ease of deployment as required. During both refloating attempts, when the vessel was broadside to the current, the tugs were not effectively used to position the Alcorin order to counter the effect of the flood tide and to facilitate safe entry into the channel. B 2.3 Pilot Relief In the minutes and hours that follow a grounding or other major incident in pilotage waters, a pilot will normally assist the master with the execution of various operational duties. In an emergency situation, such as a grounding, time is of the essence. The duration of the emergency can extend into many hours or days, before a refloating attempt is successful. In this context, factors affecting crew/pilot performance include: the number of hours worked, the ability to get regular and uninterrupted sleep, and the exposure to stressful conditions, both mental and physical.43 A refloating attempt is an extremely demanding undertaking. One of the elements necessary for a successful mission is the pilot's performance and his/her ability to retain full concentration. In order to ensure such a level of concentration, the pilot needs to be well rested and, ideally, emotionally removed from the occurrence. While there are provisions made to contact pilots in emergency situations and offer relief, the decision to request relief or assistance rests with the occurrence pilot. In this occurrence, the pilot was offered relief, but declined. At the time of the failed refloating, he had been on board for some 38hours, during about 30of which he was actively involved in operations. While the pilot had a cabin at his disposal and had the opportunity to sleep, the quality of his sleep may not have been ideal, as there had been a flurry of activity subsequent to the initial grounding and the developing cracks in the hull produced loud reverberations during the night. Deterioration in performance due to fatigue is characterized by, among other things, a slower reaction time, errors, false responses, and decreased vigilance. From an operational perspective, this can lead to compromised attention, limited situational awareness, and judgement processes clouded by a failure to reliably detect, appreciate, and respond to events in a timely manner.44 This occurrence embodies several of these indicators. The interaction between the salvage master and the pilot, moments before the second grounding, is one example. While backing off the shoal, the salvage master asked the pilot if the vessel was in safe water. The pilot responded in the affirmative (which was technically true), but did not advise the salvage master that the vessel was quickly running into danger again. Additionally, no attempt was made to stem, or stern, the current and wind once the vessel was afloat. This is not an isolated occurrence. For example, in 1997 the bulk carrier Venus grounded near Bcancour, Quebec. The pilot, under contract to LPA, had elected neither to seek relief nor to request an additional pilot to share the workload, and had remained on board for an extended period. Both the Venus and the Alcorran aground for a second time after refloating. While not necessarily causal, degradation in pilot performance due to fatigue has been identified as a factor in second groundings. At the centre of the LPA pilot relief system (subsequent to an occurrence) is the notion that a pilot can carry out a self-assessment with respect to fatigue and emotional state. However, individuals do not reliably estimate their level of alertness and performance.45 The insidious nature of fatigue and its impact on decision making and judgment has been highlighted in previous TSB reports.46 Further, stress associated with having been involved in an occurrence may have an impact on a pilot's ability to perform his/her duties. It has long been known that stress can induce certain types of error. Finally, human nature and professional pride can hinder any objective self-assessment of a pilot's need to be relieved or assisted. While the need for relieving a pilot involved in an occurrence has been recognized by the Great Lakes Pilotage Authority,47 LPA does not require such relief, though they do recognize the need to relieve a pilot under normal conditions when a voyage is extended due to a slow ship, or in winter conditions. In the absence of clear criteria regarding relief of pilots subsequent to an occurrence, a pilot is placed in the difficult position of making a decision on whether to request relief or assistance. Under the circumstances, a pilot may not be best suited to make this decision, a decision which can have an impact on navigational safety. B 3.0 Conclusions B3.1 Findings as to Causes and Contributing Factors During both refloating attempts of the Alcor, the tugs were not used to advantage to counter the effects of wind and tide. This resulted in the vessel grounding a second time on 10 November 1999, and narrowly missing a third grounding on 05 December 1999. Initial response to the grounding emergency was less than adequate. The lack of timeliness, and misjudgement of the resources needed to free the vessel, diminished the chances that the vessel would be refloated successfully before damage was incurred. B 3.2 Findings as to Risk lack of control, leading to confusion and uncoordinated activities among government departments and agencies and commercial enterprises; and inappropriate allocation and ineffective use of resources, be they personnel, technology, or equipment. lack of control, leading to confusion and uncoordinated activities among government departments and agencies and commercial enterprises; and inappropriate allocation and ineffective use of resources, be they personnel, technology, or equipment. LPA has no structured approach that would prepare pilots to help masters in making informed decisions regarding emergency response. During salvage operations, communication among members of the bridge team regarding navigation of the Alcorwas limited and casual, effective use was not made of all available navigational equipment, and the working relationship was fragmented. The less-than-effective application of bridge resource management principles increased the risk of an accident by becoming a weak point in a system prone to single-point failure. Fatigue and the stress of emotional involvement in an occurrence can preclude an accurate self-assessment of pilot performance and of the need for relief or assistance, increasing the chances of another accident if a pilot stays on board for an extended period. A coordinated and structured approach to navigation-related emergencies was not used, and this precluded an objective assessment of the emergency response. B 3.3 Other Findings The 24-hour delay before government intervention (pursuant to the Navigable Waters Protection Act) may, under certain circumstances, increase risk to the safe transit of vessels. C 2.0 Analysis C2.1 Limitations Imposed by Navigational Practices No prior arrangement was made among pilots of various vessels at anchor as to their order of departure, nor did VTS establish the order for vessel departure. Instead, each vessel made a decision in isolation and commenced weighing anchor; the Eternity first, followed by the CanmarPride. As the Eternity was underway, she was required to keep clear of vessels at anchor. Although the pilot of the Eternity, the farthest vessel at anchor, was aware that he would have to pass other vessels at anchor, he did not closely monitor these vessels nor did he communicate with them to arrange safe transit. The pilot of the Eternity was aware that the CanmarPride had begun weighing anchor. In fact, it was the pilot on board the CanmarPride who twice raised concern about the developing situation. The pilot of the Eternitydid not establish the precise position of the CanmarPride, and close monitoring of the situation would have better enabled him to recognize, in sufficient time, that the CanmarPride was anchored close to the channel. A review of the chart indicates that the channel in the vicinity of CanmarPride anchorage is some 0.9mile wide. Hence, as the Eternity approached the CanmarPride, the pilot on the former had a number of options available to him: to reduce the vessel's speed, or to pass either ahead or astern of the CanmarPride. The Eternity's pilot opted to stay on the recommended track (marked on the chart), maintain her speed of 12knots and pass ahead of the vessel. Once committed to the port alteration, and with the wind and current on the vessel's port quarter, the Eternitypilot was left with little alternative but to take last-minute emergency action; sufficient allowance had not been made for the vessel's set given the position of the CanmarPride. This action, in conjunction with the emergency action initiated by the CanmarPridepilot (going astern on the engine), barely extricated the vessels from a collision; passing clearance was some 30m. C 2.1.1 Issues Affecting Quality of VTS Because there is no radar coverage of this area, VTS had to rely on information provided from vessels (by either the pilots or shipboard personnel) to generate a traffic image. To achieve this, VTS procedures call for communicating positions of vessels at anchor. In this instance, the anchored position of the CanmarPridewas reported as being at Pointe Saint-Jean anchorage. No range and bearing were given, nor was it requested by VTS. As the vessel was anchored close to the recommended track, she posed a potential threat to transiting traffic, given the circumstances which existed at the time of this occurrence. At no time did the CanmarPride broadcast a SECURIT message to advise other vessels of this position. The absence of this information precluded VTS and vessels in the area from having a comprehensive overview of the traffic in the area and traffic-influencing factors - criteria essential for the safety of vessels operating within the VTS area. Further, the quality of VTS accident prevention measures depends on the system's capability of detecting a developing dangerous situation and giving timely warning of such dangers.51 Precise information on anchorage positions enables the VTS officer to better understand the traffic situation and to more fully inform other vessels navigating in the area. Furthermore, precise VTS information allows vessels to better appreciate the risks associated with traffic in the area, thereby enhancing the safe conduct of vessels. C 2.2 Closing and Re-opening of Channel C 2.2.1 VTS Control and Direction C 2.2.1.1 Normal Operating Conditions The CSA requires that a vessel obtain a traffic clearance before departure. There is no intention on the part of CCG to attempt to navigate or manoeuvre ships from a shore station. Information provided to a vessel is intended to assist in the safe conduct of that vessel. Consequently, pilots make navigational arrangements between themselves and keep VTS apprised, so that pertinent information can be disseminated to other vessels. Under normal operating conditions, the system works well and the safety of vessels navigating in the VTS area is not compromised. Direct arrangements between pilots, therefore, have become an accepted group norm. C2.2.1.2 VTS Procedures for Closing and Re-opening of Channels A channel closure could result in placing a number of vessels at anchor or berthed, awaiting transit. Unlike normal operations, direction from the VTS (a central coordinating body) becomes essential for an orderly flow of traffic. This occurrence shows what can happen when this system becomes inoperative; the safety of vessels operating in the area is compromised and the risk to the environment is increased. C 2.2.2 Involvement of Participants VTS operations were influenced by the salvage operations; closing and re-opening the Traverse du Nord was one action under consideration. In this instance, the salvage plan was incomplete, in that it did not involve VTS, LPA or the Pilot Corporation at its inception. VTS was not apprised during the planning, development and execution of the refloating manoeuvre, and no specific directive was issued to VTS that would have helped an orderly flow of traffic. C 2.2.3 Notice to Shipping (NOTSHIP) The NOTSHIP (closure of channel) lacked specifics, in that it did not indicate that the channel was closed to other traffic until further notice. Consequently, once the Alcorwas refloated, each pilot unilaterally initiated action to transit the channel. C 2.2.4 Coordination between TC and Pilots Although TC officials had officially closed the channel to all traffic, the pilot(s) aboard the Alcor, with the tacit acquiescence of TC officials, made arrangements (with pilots of other vessels) to permit other vessels to transit the area (at1740). This could have been interpreted by other vessels to mean that the channel was reopened. This underscores the need for centralized control and direction over channel status. C 2.2.5 VTS Control and Authority Gradient VTS did not exercise authority to direct traffic, but instead acquiesced to the pilots' actions. A pilot has many years' seagoing experience and interacts with a VTS officer who may have little or no sea experience. The resulting authority gradient generates a barrier that makes the VTS officer less prone to exercise authority, even in extenuating circumstances, effectively undermining VTS procedures. The exercising of authority by VTS at this stage would have helped establish an orderly flow of traffic. On the other hand, lack of clear directives to VTS from either the pilots or TC authorities on the Alcor meant that MCTS officials were operating in a void, with no clear direction. Given the complexity of navigational considerations, the coordination required, and the impact of the authority gradient, the participation of personnel well versed in the area of operation and the application of VTS operating procedures could help prevent the development of an authority gradient. C2.3 Effectiveness and Impact of Communication information provided by VTS to the vessels was limited to the opposing traffic; communication between vessels was rudimentary and did not reflect the added safeguards required in extenuating circumstances; no SECURIT message was broadcast by vessels departing anchorage; and the anchorage position of the CanmarPridewas close to traffic lanes and potentially an impediment for the transiting vessels. This was not reported to VTS or broadcast by the CanmarPride. A number of TSB investigations have highlighted the fact that accidents are often the product of ineffective, incomplete, untimely, or misunderstood communications.52 A 1995 TSB study found that lack, or misunderstanding, of communication were significant factors identified in some 18% of the marine occurrences involving human factors.53 The Board was concerned that unclear communications and delays in expressing concerns continued to compromise the safety of lives, vessels, and the environment.54 This occurrence again highlights the importance of clear, complete and well-understood communications, be they among bridge team members, or between vessels, or between vessels and VTS. Further, it also highlights the need for effective control and coordination within the VTS operating area. Without it, vessels will operate in a void and navigation personnel will continue to make assumptions to the detriment of transportation safety. Many accidents in the past, in particular, collisions, have been attributed to decisions based on assumptions that ultimately proved to be erroneous.55 Risks associated with incomplete information have been highlighted in Rule7, Risk Of Collision of the Collision Regulations, which reads, in part: assumptions shall not be made on the basis of scanty information The Chief Inspector of Marine Accidents (Great Britain) recently wrote, in many instances a contributory factor to whatever the eventual incident was, involved someone assuming something was going to happen, or had been done.56 In this occurrence, the following assumptions were made: The pilot on the Eternitydid not verify that the channel was unimpeded. Information provided by VTS did not contradict this. The pilot on the CanmarPride assumed he was to be the first to proceed downriver once the channel was re-opened. Consequently, he saw nothing inopportune with his anchorage position. Conditions that led to this belief included the perception that his vessel was the closest to the entrance of the Traverse du Nord, and that his vessel was the fastest of the group waiting to proceed downriver. VTS assumed that the CanmarPridewas anchored in the manner commonly employed for the Pointe Saint-Jean anchorage. While it is recognized that erroneous assumptions can contribute to an occurrence, the availability and precision of information shapes the nature and the number of assumptions being made. The impact of incomplete or ambiguous information can be minimized through timely sharing of precise and complete information. Safety is dependant upon, among other things, the level of shared situational awareness amongst the individuals piloting vessels in the channel. Situational awareness can be thought of as the mental model that an individual has of a given situation and time. Mental models develop from information related to the immediate situation and environment (such as location, speed, and presence of hazards) and information gained from education, training, and experience. In the absence of a complete set of cues for a given situation, fragmentary information may be combined with mental expectations and integrated (in the form of assumptions) into the mental model. In such situations, it is possible for different individuals to develop divergent models of their surroundings, even though they had the same information as a starting point. Precise and complete communications are essential to generate a comprehensive overview of traffic and the factors influencing it. The resulting shared situational awareness increases the probability that a developing dangerous situation is recognized in a timely manner. C 2.4 Marine Occurrence Reporting Timely collection of occurrence information is an essential component of any safety system. Such timely reporting ensures that the relevant authorities are quickly apprised, so that search and rescue, pollution prevention, inspection agencies, and other organizations can be dispatched to mitigate risk to personnel, property, and the environment. Furthermore, it permits quality accident investigation action and provides a knowledge base through which trends can be analysed, deficiencies identified, and recommendations for change brought forth. At a regional level, pilotage authority reporting requirements mandate that pilots submit incident reports so that the local authority can, where necessary, undertake an investigation with respect to policy, procedures, and practices. Reporting also provides the authority with a risk assessment tool to identify latent safety deficiencies before they lead to a major occurrence. Neither pilot properly advised VTS or any other agency, including the LPA, of the near collision. When the LPA became aware of the near collision, after some preliminary inquiries it decided not to investigate further. Human nature being what it is, a near miss quickly gets downgraded into 'part of the job' and is forgotten.57 This decision had the effect of obscuring (to the Authority) the safety sensitive information of vessels operating in pilotage waters. In Canada, the TSB maintains a national occurrence database and information is available to the public both nationally and internationally. The prime purpose of a pilotage service is safety. Compulsory pilotage areas are established for the benefit of the community to protect the environment, the waterway, and the port infrastructure from marine accidents. There is an expectation that all incidents will be properly reported and investigated. This will allow for the identification of safety deficiencies and commensurate safety action, thereby advancing transportation safety. C 3.0 Conclusions C3.1 Findings as to Causes and Contributing Factors The navigation team aboard the Eternity did not appreciate, in a timely fashion, that a risk of collision with the CanmarPridewas developing, and they did not initiate communication with the former to arrange a safe passage. The lack of coordination between VTS, the pilots, and TC officials, combined with incomplete and imprecise communication, led to divergent interpretations of unfolding events, both ashore and afloat, resulting in confusion and the uncoordinated reopening of the channel. The anchorage position of the CanmarPride,close to traffic lanes and potentially an impediment to transiting vessels, was not communicated to VTS or to other vessels by way of direct, two-way communications or by a SECURIT message, thus depriving other vessels of a vital navigation cue. A lack of procedural integrity by VTS operators, and the impact of an authority gradient between the pilots and the VTS operators, resulted in a loss of direction and control with respect to the orderly flow of traffic. C 3.2 Findings as to Risk LPA did not fully investigate the events surrounding the near collision and were unable to avail themselves of safety-sensitive information. C 3.3 Other Findings The near collision was not reported in a timely manner by either the Eternity or the CanmarPride.The quality of information and timely identification of safety deficiencies are compromised when transportation occurrences are reported late, or not at all. D 2.0 Analysis D2.1 Bulk Carrier Inspection and Enhanced Surveys D 2.1.1 Hull Failure Sequence The grounded after end of the Alcorwas supported while buoyant support of the forward end fluctuated throughout the diurnal tidal ranges. Consequently, the vessel was repeatedly subjected to very high bending moments. Because two of the forward double bottom tanks were breached during the initial grounding, the forward end was only partially buoyant and the resulting downward deflections of the bow at each low tide caused the hull to take up a hogged attitude. While in the hogged condition, the upper member (main deck and its longitudinals) of the hull girder was in tension and the lower member (bottom shell and inner bottom) was in compression (seeFigure8). For vessel construction, the midship section modulus, configuration, and grade(s) selection of steel are designed to ensure that the as-built intact structure will, with safety margins, withstand loads likely to be imposed on a vessel in the prescribed cargo and ballast loading conditions encountered when afloat in normal service. However, in the event of grounding, when total buoyant support is reduced and longitudinal distribution is much less uniform than when free-floating, bending stresses imposed on a hull can exceed the safety margin ensured by the approved maximum design figure. Ultrasonic inspection showed that the average reductions in thickness of the main deck plating, the most recently renewed longitudinals, and the sheer strake and upper side shell were within regulatory limits before replacement was required. However, some of the average or localized wastage and corrosion pitting in many of the original deck longitudinals, together with web frames and lower sloping bottom plating in the upper wing water ballast tanks, exceeded the maximum 25%and 30%allowable reduction margins before replacement was necessary. The grounded hull was subjected to bending stresses beyond normal design criteria. The ability of the vessel to resist longitudinal bending and wracking stresses, however, was reduced due to corrosion wastage, localized pitting and damage to parts of the upper hull structure. Because of the river bottom profile at this location and the position and loaded condition of the Alcor, it is highly likely that hull failure would have eventually occurred, regardless of hull condition. Nonetheless, deterioration of the upper hull structure contributed, in part, to the rapidity of the hull failure. D 2.1.2 Quality of Repairs Extensive repairs carried out at Shanghai, People's Republic of China, in 1997 were concentrated on the upper wing water ballast tanks. Several sections of main deck longitudinals were renewed, as were the repair or partial replacement of transverse webs, end bulkheads, and main deck and wing tank sloping bottom plating. After the Alcorwas damaged, an opportunity to conduct an unscheduled (and independent) inspection was possible. The ends of the exposed longitudinals in way of the main deck fractures showed fillet-welded connections to the main deck plating with irregular throat sizes and leg lengths. There was also grooving and undercutting of the deck plating and a lack of penetration of some of these welds. A further 19fractures were found in the main deck longitudinals at other locations. Several of these fractures were in way of butt joints, where the weld had incurred preferential corrosion; others were in way of heavily-corroded and locally-thinned metal of original longitudinals. The varying quality of repairs has been previously underscored.58 Coincidentally, the sister ship Cheetah also had certain sections of the deck longitudinals recently renewed. The PSC inspection conducted on 07April2000 revealed that fillet welds, joining the old with the new, were the origin of cracks found on deck. Quality control during vessel construction is an essential part of ensuring a safe vessel. The same rigorous standards for quality control used during construction should also be applied to repairs, especially major repairs to a vessel's principal structural elements. D 2.1.3 Quality of Inspections The internal structure of port and starboard upper wing water ballast tanks 2 and 3 showed extensive and active corrosion when inspected by TSB investigators in November 1999. Routine annual class surveys in 1998 and 1999 both acknowledged the absence of cathodic protection systems or protective coatings for these areas. The 1999 survey noted that the internal structure of all upper wing water ballast tanks was in satisfactory condition. It was also recorded that there was not found substantially corroded areas. Although it is inherently difficult to measure the success of the ESP, some improvement in safety has resulted. However, unsatisfactory conditions remain. In at least four highly publicized accidents since December 1999, hull failure due to structural inadequacies was suspected. Each vessel was subject to the ESP, and was duly certified and classed: Erika, broke and sank off France in December1999. LeaderL, broke and sank in the Atlantic Ocean in March2000. Treasure, broke and sank off South Africa in June2000. and LevoliSun,broke and sank off France in October2000. With the Alcor, there was a significant difference between the observed condition of the upper wing ballast tanks in November1999 (after the grounding), and the condition reported in the January1999 routine annual survey. It would appear unlikely that such a marked deterioration could occur during the 11-month period between the two surveys, since these tanks routinely were kept dry, and hold3 was used as the principal means of ballasting the vessel. This occurrence, and the four mentioned above, reveal concerns with the quality control of inspections conducted under existing ESP procedures. As a foreign bulk carrier entering a Canadian port, the 23-year-old Alcorwas subject to inspection by TCMS in accordance with PSC and the Bulk Carrier Inspection Programme requirements. Such an inspection was scheduled for 10 November 1999, at Trois-Rivires, Quebec. D 3.0 Conclusions D3.1 Findings as to Causes and Contributing Factors The grounded vessel was subjected to bending and racking stresses that exceeded normal operational design criteria. However, the deteriorated condition of parts of the Alcorupper hull structure contributed to the rapidity of hull failure. D 3.2 Findings as to Risk The quality of inspections conducted under the existing ESP is not consistently held to the standard required by the program. Consequently, vessels which may be unseaworthy continue operating, thus compromising the safety of these vessels and subjecting personnel, property, and the environment to unacceptable risk. The quality of welding repairs in way of the renewed main deck longitudinals was less than satisfactory. E 4.0 Safety Action E 4.1 Action Taken E 4.1.1 Marine Communication and Traffic Services Subsequent to the near collision of 05 December 1999 between the vessels CanmarPride and Eternity, Marine Communications and Traffic Services (MCTS) Qubec undertook an internal review. The results of this review included the following findings: Information transmitted to and from VTS was, at times, unclear and did not conform to the accepted normalized lexicon. Several internal procedures were not followed, in particular noting a vessel's exact anchorage position (CanmarPride). Clear and unequivocal authorization to depart anchorage was never given by VTS. Authorizations were tacit and in response to declarations made by the respective pilots as they were departing - a subtle but fundamental difference that contradicts VTS standing procedures and the CSA. Downbound traffic was not managed once the Alcor left the Traverse du Nord, in particular the staging of departures to ensure safe and efficient navigation within the channel. Additionally, the internal MCTS review identified the use of unmonitored VHF channels by pilots to exchange information that should normally be communicated on monitored channels. The workload of the VTS officer was unduly increased due to their unofficial monitoring of these communications and then having to ask the pilots to confirm, on monitored channels, actions which were now known to VTS but not known to other river traffic. The lack of recordings for these unmonitored channels renders post-incident analysis difficult, if not impossible, and precludes confirmation of this practice. Actions taken by MCTS after the groundings and the near-collision include: documenting all information and parameters of the events relevant to these incidents, with a view to developing a simulation scenario for training purposes; increasing internal quality control measures with respect to procedural integrity and quality of decision making by VTS officers and team leaders; reviewing operating procedures within the VTS station with a view to increasing the active participation of VTS officers and improving traffic management; and meeting with all VTS team leaders to instill the importance of quality decision making in all aspects of their service. E4.1.2 Transport Canada and Fisheries and Oceans Canada Subsequent to the Alcorgroundings and refloating, regional representatives of Transport Canada and Fisheries and Oceans Canada held a combined review and agreed upon common action: to identify a go team for initial response; to develop and ratify a memorandum of understanding (MOU) between the two ministries for environmental response; to ensure an annual review of the environmental response MOU; and to create a permanent inter-departmental working group. Additionally, the review, which is still underway, will consider the need for TC to develop a risk evaluation methodology for marine incidents. An MOU was signed with Fisheries and Oceans Canada on 19October2000. The MOU defines the role responsibilities of the TC marine safety surveyor as a representative of CCG in the event of marine pollution from a vessel, specifically in circumstances where the surveyor boards the vessel before (and until) the CCG representative arrives on board. The TSB issued Marine Safety Information Letter (MSI04/01) to TC on 24May2001. This information letter points out the shortcomings of working in an ad hoc fashion and stresses the advantages of a risk management model for navigation-related emergencies, including groundings in high tidal amplitude waters. Fisheries and Oceans Canada took the following action: An MOU was signed with TC on 19October2000 (sameasabove). Local area (grounding environs) sounding criteria have since been established as guidance for salvors. A list of potential local contractors that meet the criteria was also developed. E 4.1.3 Transport Canada The three sister ships of the Alcor, as well as all bulk carriers constructed at the same shipyard for the three-year period extending one year before and after the Alcor's construction, were identified as SPIs and communicated to signatories of the Paris and Tokyo PSC Protocols. E 4.1.4 Pilot Corporation E 4.1.4.1 Training Since 2001, at the insistence of the Pilot Corporation, the ship-handling course has been made available to ClassC pilots. As of August2002, all ClassC pilots have taken this course. E 4.1.5 Laurentian Pilotage Authority E 4.1.5.1 Pilot Relief The TSB issued Marine Safety Information Letter MSI05/01 to the Laurentian Pilotage Authority (LPA) on 18May2001. This information letter points out that, in the absence of clear criteria regarding relief of pilots subsequent to an emergency, an occurrence pilot is placed in a difficult position of making a decision. It goes on to emphasize that, under the given circumstances, a pilot may not be the best person suited to make the decision - a decision that may have a significant impact on the safe navigation of the vessel. As yet, no formal program has been instituted by LPA to address relief of pilots subsequent to special circumstances such as a grounding or other emergency. E 4.1.5.2 Pilot Training The TSB issued a Marine Safety Information Letter (MSI06/01) to the LPA on 24May2001. This information letter expressed concern that the current training and experience requirements for C-1pilots, and the methodology used to evaluate them, are such that pilots may not be sufficiently prepared to pilot larger vessels. Further, it goes on to highlight that there is no effective competency-based training and evaluation program to help ensure that candidates possess the required abilities and attain a level of competency commensurate with the appropriate class and sub-class of licence. Following recommendation9 of the 1999 Canadian Transportation Agency's (CTA) Review of Pilotage Issues, Report to the Minister of Transport, the pilotage authorities have begun the process of developing a system for assessing pilots' competence and quality of service. E 4.1.5.3 Risk Assessment Following Recommendation1 of the 1999 CTA pilotage review report, the LPA, in conjunction with the other pilotage authorities, has since developed a risk evaluation tool. For 2002, the objective was to use this new tool to evaluate three priority issues; 1- compulsory pilotage limits in so far as vessel size within the Laurentian region, 2-the use of docking pilots in District2, and 3- double pilotage requirements in Districts1 and2. The work is ongoing. E4.1.5.4 Regulatory Change - Shipping Casualty Reporting E 4.1.6 Sister Ship Operators Subsequent to Marine Safety Information Letter (MSI05/99), the manager of two sister ships, the Cheetah and the Lynx, required a protective latch to be installed on the steering mode selector (SMS) switch of the Hokushin helm units of both. The protective latch positively locks the switch in the HAND mode and must be lifted in order to turn the SMS switch to another position. Subsequent to the numerous structural defects found during the PSC inspection of the Cheetahat Sept-les on 07April2000, the vessel was detained for repairs. Completion of repairs required 10days. E 4.1.7 Russian Maritime Registry of Shipping Since the Alcoroccurrence, the Russian Maritime Registry of Shipping has implemented guidelines for surveys of vessels more than 20yearsold. Also, a working group, based out of St.Petersburg, Russia, has been established to serve a quality control function for surveys of bulk carriers and tankers under the ESP. E 4.2 Action Required E 4.2.1 Emergency Preparedness When responding to accidents, such as groundings involving large vessels, the situation can be complex, involving several different requirements, agencies, and personnel. Time also plays a critical role in determining what corrective action can be taken. Decisions have to be taken quickly and, with a measure of uncertainty as logistics associated with the mobilizing of people and equipment to accident sites may require significant effort. Further, solutions to problems must be carefully considered by all parties as the actions taken to resolve the situation do, in themselves, present their own risks that have to be properly assessed and managed in a dynamic environment. Instances are on record where incomplete/improper risk assessment of vessels involved in an accident has led to an escalation of the incident. Examples include, among others, the TorreyCanyon(1967), the ExxonValdez (1989), the AmocoCadiz (1978), the SeaEmpress (1996). Similar safety deficiencies have been repeated in the accident involving Alcorand the subsequent events. The shortcomings identified in the Alcor occurrence were: inadequate response to the initial grounding contributing to the escalation of the incident; less-than-optimal use of tugs during salvage operations, culminating in the vessel running aground a second time and sustaining extensive damage; all of the key parties were not involved in the planning and development of the salvage plan, resulting in an uncoordinated approach to refloat the vessel and the near collision of the tanker Eternity and the container vessel CanmarPride; the working relationship of the bridge team during salvage operations was fragmented and uncoordinated, with the vessel almost grounding for a third time; there was no contingency plan for navigation-related emergencies, leading to an uncoordinated approach to handing the emergency, and the lack of a contingency plan precluded objective assessment by government officials of the timeliness and appropriateness of the emergency response. Following the grounding of a vessel, it is incumbent upon the owner to take timely and appropriate action to respond to the situation and initiate remedial action. Various governmental organizations are on scene, each with complementary and interlaced mandates. TC has the general superintendence of all matters relating to the safe operation of ships, salvage and, subject to the Canadian Transportation Accident Investigation and Safety Board Act, shipping casualties. Fisheries and Oceans Canada has the general superintendence of all matters relating to the navigation system, wrecks and receivers of wrecks, and for responding to discharges of pollutants from vessels. Pilotage Authorities are involved in that they provide assistance to the vessel's navigation team. In the aggregate, these government organizations have a responsibility and an accountability to ensure that actions by the owner are correct and address the risks. Therefore, it is essential that they be fully prepared. In the marine environment, the Canadian Coast Guard (CCG) has contingency plans for responding to maritime search and rescue (SAR) and discharge of pollutants by vessels. However, no formal structured management system, together with an overarching contingency planning, exists for other navigation-related emergencies in Canadian waters. During emergencies, such as a grounding, various governmental representatives arrive on scene, principally TC and Fisheries and Oceans Canada, and they assume an observer/advisor status. However, as an emergency escalates many agencies/ departments at all levels of government (federal, provincial, local) as well as commercial interests could be involved. The role of an observer/advisor is to assess timeliness and appropriateness of the emergency response and, where deemed appropriate, give directions in relation to the emergency response.59 However, there is no performance criterion against which the emergency response can be measured. Instead, there is heavy reliance placed on personnel expertise and experiences. Appropriate tools are not provided to facilitate the objective assessment against TC's or Fisheries and Oceans Canada's expectations of the timeliness and appropriateness of measures/actions taken by the owner's representatives and other entities. Without a formal structured management system, together with an overarching contingency planning for navigation-related emergencies, the observer/advisor's effectiveness is limited to the individual's initiative and that person's ability to fully grasp the complexities and to influence or, at some point, to direct or to stop the action to be taken because the desired outcomes cannot be reasonably achieved. The owner's need to be prepared for an emergency and to take relevant action on board has been recognized in Article8 of the International Safety Management Code (ISMCode). The Code requires companies to establish procedures/plans to identify and respond to, among others, navigation-related emergencies such as a grounding. In an emergency situation, there are several factors which are beyond the knowledge/expertise of the ship's master/owners; e.g., local conditions/environment, availability of resources, etc. There is, therefore, a need for the authorities to be prepared to assess the adequacy of responses to emergency situations and to take appropriate action to facilitate the implementation of the plan. A salvage operation, by its very nature, is extremely complex and fluid; success depends upon the thoroughness of the plan and the effectiveness of its execution. The dynamic nature of salvage operations often requires ad hoc problem solving and decision making, a situation conducive to increased error. A structured approach, therefore, provides a framework around which informed decisions can be made. Such an emergency management approach by TC and CCG helps ensure that the vessel owners' response to deal with the emergency at various stages of its development is effective and appropriate under the circumstances. Further, it would ensure the owner's response plan identifies all risk and considers all risk mitigating options. Such an integrated approach would foster prudent and effective understanding, decision making and communication of the measures taken to resolve the emergency and permit continuous evaluation of its effectiveness. The Board recognizes that emergency response management structures and risk-based decision-making models are used in response to specific marine emergency situations that do not include response to navigation-related emergencies. Further, noting the complementary mandates of TC and CCG to foster the safety of vessels and to protect the marine environment and, acknowledging the important role of pilotage authorities in providing valuable information on the operation of ships in pilotage waters, the Board believes that a planned and coordinated approach is necessary to deal with navigation-related emergencies in Canadian waters while supporting the vessel owners' efforts to deal with an occurrence. The Board, therefore, recommends that: The Department of Transport, the Department of Fisheries and Oceans, and Canadian pilotage authorities, in consultation with marine interests, develop, implement and exercise contingency plans to ensure that risks associated with navigation-related emergencies are adequately addressed. M03-03 Assessment/Reassessment Rating: Satisfactory Intent E 4.3 Safety Concern E 4.3.1 Pilot Performance The critical role of pilots in ensuring the safe passage of vessels is well understood. Pilots work on a variety of ships with diverse handling characteristics and, at a critical phase of a ship's voyage, are introduced to bridge teams with a wide range of cultural and linguistic aspects. The challenge, therefore, in such a dynamic operating environment, is to ensure that pilots have the necessary competencies and support to be able to carry out their duties in the wide range of circumstances they encounter. A number of TSB investigations involving vessels under the conduct of a pilot have revealed safety deficiencies for which recommendations were made.60 The safety issues addressed factors that affect pilot performance. These included the following: pilot/master information exchange; pilot performance degradation due to fatigue; pilot skill upgrading, training and training validation; pilot bridge resource management training and practices; and pilot fitness for duty. The investigations also revealed that a systematic approach to fatigue management and periodic audits to evaluate pilot proficiency and skills are usually absent from pilotage organization regimes. Thus, in 1999, following the investigation of the grounding of the bulk carrier Raven Arrow, the Board recommended that the pilotage authorities develop and implement a safety management quality assurance system.61 This recommendation was issued on the heels of a similar recommendation issued by the CTA, which called for pilotage authorities to develop and implement a fair and reasonable system for assessing pilots' competence and quality of service.62 In response to the two recommendations, the Minister of Transport tasked the pilotage authorities to develop a quality assurance system in accordance with the needs and characteristics of their respective regions.63 The pilotage authorities subsequently introduced a methodology for risk-based decision-making (pilotage risk management methodology) for use by pilotage authorities to ensure the efficiency, viability, and safety of the Canadian pilotage system and respond to the legitimate needs and expectations of all its users. The Board commends the actions taken as a positive step toward furthering safety of vessels operating in the Canadian pilotage waters. However, the Board notes that the application of pilotage risk management methodology is limited to supporting decisions for designated compulsory pilotage areas, the size and type of vessels subject to compulsory pilotage, denying requests for waivers, and the requirement for double pilotage in designated pilotage areas. As a consequence, additional improvements to the safety of operation of a vessel in pilotage waters that may be associated with other pilotage conditions, practices, and procedures cannot flow naturally from a systemic analysis. The TSB is aware that among some of the pilotage authorities: the extent of training for preparing pilots for navigation-related emergencies and to respond to non-routine events is limited; attendance at certain pilotage training courses is dependent upon seniority; there is no requirement for competency audits of pilots after training; training validation is not carried out to ensure that the training given to pilots is effectively transferred from the classroom to the operational front; duty hours for pilots vary, as do the requirements for the need for a second pilot; and no formal guideline or procedure is in place concerning the relief of pilots involved in an occurrence. Implementation of a risk-based methodology approach to address all pilotage issues, be they regulatory, contractual, or operational, would ensure that risk reduction is a prime consideration in the operational environment. In 1998, a study entitled Modernization of the Pilotage Certification Processin the Laurentian Pilotage Region64 was undertaken to examine the modernization of the pilotage certification process for shipboard personnel (the study did not look at the pilot licensing process) and highlighted several pilot performance-related issues. These issues included a competency-based training and validation program, performance-based testing and an infrastructure necessary for program delivery. The Board also notes that the pilot performance-related issues identified by this study could be applied equally well, but remain, for the most part, unaddressed within the pilot licensing process. Effective safety management systems enable organizations to identify safety deficiencies and evaluate the associated risks so that corrective action can be taken or the risk minimized before accidents occur. Given that the current application of the pilotage authorities' risk management methodology is limited and that the pilot licensing process has yet to fully address key performance-related issues, the Board is concerned that residual risks continue to exist and may compromise overall safety of other pilotage operations and performance. This report concludes the TSB's investigation into this occurrence. Consequently, the Board authorized the release of this report on 28May2003.